Standard Deviation

[WilliamLP]

One other interesting result from pure math is that if you multiply a number of trials by n, you divide the standard deviation by the square root of n.

So as an approximation, let's say your SD is 15BB over 100 hands, and your mean is +3BB/100 hands.

Then over 10,000 hands, your SD would be 15 / (square root of 100) or 1.5.

So the final results after 10,000 hands could look like:
<0BB with 2.3% chance
0-1.5BB with 12.6% chance
1.5-3BB with 34.2% chance
3-4.5BB with 34.2% chance
4.5-6BB with 12.6% chance
>6BB with 2.3% chance

So math supports the idea that your results over 10,000 hands could very realistically be all over the map.